Diagnose location-sorted recommender causing revenue drop
Company: Uber
Role: Data Scientist
Category: Machine Learning
Difficulty: medium
Interview Round: Onsite
Overview: Diagnose why a food-delivery recommender ranked mainly by distance raised add-to-cart rate but lowered revenue per session, then design a replacement. Cover success metrics, experiments separating causal from compositional effects, safe exploration, and rollback criteria.
Read the full Uber Data Scientist interview experience this question came from